Identification of Tomato Leaf Diseases Based on Improved Lightweight Convolutional Neural Networks MobileNetV3

doi:10.12133/j.smartag.SA202202003

Abstract

Abstract:

Timely detection and treatment of tomato diseases can effectively improve the quality and yield of tomato. In order to realize the real-time and non-destructive detection of tomato diseases, a tomato leaf disease classification and recognition method based on improved MobileNetV3 was proposed in this study. Firstly, the lightweight convolutional neural network MobileNetV3 was used for transfer learning on the image net data set. The network was initialized according to the weight of the pre training model, so as to realize the transfer and fine adjustment of large-scale shared parameters of the model. The training method of transfer learning could effectively alleviate the problem of model over fitting caused by insufficient data, realized the accurate classification of tomato leaf diseases in a small number of samples, and saved the time cost of network training. Under the same experimental conditions, compared with the three standard deep convolution network models of VGG16, ResNet50 and Inception-V3, the results showed that the overall performance of MobileNetV3 was the best. Next, the impact of the change of loss function and the change of data amplification mode on the identification of tomato leaf diseases were observed by using MobileNetV3 convolution network. For the test of loss value, focal loss and cross entropy function were used for comparison, and for the test of data enhancement, conventional data amplification and mixup hybrid enhancement were used for comparison. After testing, using Mixup enhancement method under focal loss function could improve the recognition accuracy of the model, and the average test recognition accuracy of 10 types of tomato diseases under Mixup hybrid enhancement and focal loss function was 94.68%. On the basis of transfer learning, continue to improve the performance of MobileNetV3 model, the dilated convolution convolution with expansion rate of 2 and 4 was introduced into convolution layer, 1×1 full connection layer after deep convolution of 5×5 was connected to form a perceptron structure in convolution layer, and GLU gating mechanism activation function was used to train the best tomato disease recognition model. The average test recognition accuracy was as high as 98.25%, the data scale of the model was 43.57 MB, and the average detection time of a single tomato disease image was only 0.27s, after ten fold cross validation, the recognition accuracy of the model was 98.25%, and the test results were stable and reliable. The experiment showed that this study could significantly improve the detection efficiency of tomato diseases and reduce the time cost of disease image detection.

Key words: tomato disease identification, convolutional neural networks, transfer learning, MobileNetV3, activation function, identification and classification

CLC Number:

TP391.41

ZHOU Qiaoli, MA Li, CAO Liying, YU Helong. Identification of Tomato Leaf Diseases Based on Improved Lightweight Convolutional Neural Networks MobileNetV3[J]. Smart Agriculture, 2022, 4(1): 47-56.

Figures/Tables 13

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Fig.8

Table 1

Fig. 9

Table 2

Fig. 10

Table 3

References 25

1	张创创, 袁帅坤, 霍克坤. 影响番茄生产效益的因素[J]. 湖北农机化, 2020(12): 33-34.
	ZHANG C, YUAN S, HUO K. Factors affecting tomato production efficiency[J]. Hubei Agricultural Mechanization, 2020(12): 33-34.
2	王建玺, 宁菲菲, 鲁书喜. 基于支持向量机的苹果叶部病害识别方法研究[J]. 山东农业科学, 2015, 47(7): 122-125, 141.
	WANG J, NING F, LU S. Study on apple leaf disease identification method based on support vector machine[J]. Shandong Agricultural Science, 2015, 47(7): 122-125, 141.
3	秦丰, 刘东霞, 孙炳达, 等. 基于图像处理技术的四种苜蓿叶部病害的识别[J]. 中国农业大学学报, 2016, 21(10): 65-75.
	QIN F, LIU D, SUN B, et al. Recognition of four different alfalfa leaf diseases based on image processing technology[J]. Journal of China Agricultural University, 2016, 21 (10): 65-75.
4	夏永泉, 李耀斌, 李晨. 基于图像处理技术的小麦叶部病害识别研究[J]. 科技通报, 2016, 32(4): 92-95.
	XIA Y, LI Y, LI C. Recognition of wheat leaf disease based on image processing technology[J]. Science and Technology Bulletin, 2016, 32(4): 92-95.
5	刘婷婷, 王婷, 胡林. 基于卷积神经网络的水稻纹枯病图像识别[J]. 中国水稻科学, 2019, 33(1): 90-94.
	LIU T, WANG T, HU L. Rhizocotonia Solani recognition algorithm based on convolution neural network[J]. China Journal of Rice Science, 2019, 33(1): 90-94.
6	WU Y. Identification of maize leaf diseases based on convolutional neural network[J]. Journal of Physics: Conference Series, 2021, 1748(3): ID 032004.
7	丁瑞, 周平. 基于卷积神经网络的典型农作物叶病害识别算法[J]. 包装学报, 2018, 10(6): 74-80.
	DING R, ZHOU P. Identification of typical crop leaf diseases based on convolutional neural network[J]. Packaging Journal, 2018, 10(6): 74-80.
8	陈桂芬, 赵姗, 曹丽英, 等. 基于迁移学习与卷积神经网络的玉米植株病害识别[J]. 智慧农业, 2019, 1(2): 34-44.
	CHEN G, ZHAO S, CAO L, et al. Corn plant disease recognition based on migration learning and convolutional neural network[J]. Smart Agriculture, 2019, 1(2): 34-44.
9	WAHEED A, MUSKAN G, GUPTA D, et al. An optimized dense convolutional neural network model for disease recognition and classification in corn leaf[J]. Computers and Electronics in Agriculture, 2020, 175: ID 105456.
10	常江, 管声启, 师红宇, 等. 基于改进生成对抗网络和MobileNetV3的带钢缺陷分类[J]. 激光与光电子学进展, 2021, 58(4): 221-226.
	CHANG J, GUAN S, SHI H, et al. Strip defect classification based on improved generative adversarial networks and MobileNetV3[J]. Laser & Optoelectronics Progress, 2021, 58(4): 221-226.
11	FAYE M, CHEN B, KANE A S. Plant disease detection with deep learning and feature extraction using plant village[J]. Journal of Computer and Communications, 2020, 8(6):10-22.
12	WANG X, ZHANG C, ZHANG S. Multiscale convolutional neural networks with attention for plant species recognition[J]. Computational Intelligence and Neuroscience, 2021, 2021: ID 5529905.
13	ZOHAIB M, SHUN F, QUOC V. Spectral images based environmental sound classification using CNN with meaningful data augmentation[J]. Applied Acoustics, 2021, 172 :ID 107581.
14	陆健强, 林佳翰, 黄仲强, 等. 基于Mixup算法和卷积神经网络的柑橘黄龙病果实识别研究[J]. 华南农业大学学报, 2021, 42(3): 94-101.
	LU J, LIN J, HUANG Z, et al. Identification of citrus fruit infected with Huanglongbing based on Mixup algorithm and convolutional neural network[J]. Journal of South China Agricultural University, 2021, 42(3): 94-101.
15	葛丽君. 混合beta分布的建模方法研究[J]. 南国博览, 2019(4): 166.
16	易强, 李成娟, 李宝清, 等. 基于改进MobileNetV1网络的野外车辆识别[J]. 工业控制计算机, 2020, 33(7): 104-107.
	YI Q, LI C, LI B, et al. Field vehicle recognition based on improved MobileNetV1 network[J]. Industrial Control Computer, 2020, 33(7): 104-107.
17	陈智超, 焦海宁, 杨杰, 等. 基于改进MobileNetV2的垃圾图像分类算法[J]. 浙江大学学报(工学版), 2021, 55(8): 1490-1499.
	CHEN Z, JIAO H, YANG J, et al. Garbage image classification algorithm based on improved MobileNetV2[J]. Journal of Zhejiang University (Engineering Edition), 2021, 55(8): 1490-1499.
18	LI X, SHEN X, ZHOU Y, et al. Classification of breast cancer histopathological images using interleaved DenseNet with SENet (IDSNet)[J]. PLoS One, 2020, 15(5): ID e0232127
19	李淼, 王敬贤, 李华龙, 等. 基于CNN和迁移学习的农作物病害识别方法研究[J]. 智慧农业, 2019, 1(3): 46-55.
	LI M, WANG J, LI H, et al. Method for crop disease identification based on CNN and transfer learning[J]. Smart Agriculture, 2019, 1(3): 46-55.
20	MOON J, HOSSAIN M B, CHON K H. AR and ARMA model order selection for time-series modeling with ImageNet classification[J]. Signal Processing, 2021, 183(10): ID 108026.
21	姚燕, 胡立坤, 郭军. 基于深度迁移网络MobileNetV3的地形识别[J]. 广西大学学报(自然科学版), 2021, 46(4): 996-1007.
	YAO Y, HU L, GUO J. Terrain recognition based on deep migration network MobileNetV3[J]. Journal of Guangxi University (Natural Science Edition), 2021, 46 (4): 996-1007.
22	AMAHAN P A, VILLARICA M V, VINLUAN A A. Technical analysis of Twitter data in preparation of prediction using multilayer perceptron algorithm[C]// Proceedings of 2021 4th International Conference on Data Science and Information Technology (DSIT 2021). New York, USA: Association for Computing Machinery, 2021: 120-124.
23	KUMAR V, SINGH R, DUA Y. Morphologically dilated convolutional neural network for hyperspectral image classification[J]. Signal Processing: Image Communication, 2022, 101: ID 116549.
24	耿海波. 基于U-Net模型的单声道唱声分离研究[D]. 乌鲁木齐: 新疆大学, 2020.
	GENG H. Research on mono singing sound separation based on U-Net model[D]. Urumqi: Xinjiang University, 2020.
25	MABUNI D, BABU S A. High accurate and a variant of k-fold cross validation technique for predicting the decision tree classifier accuracy[J]. International Journal of Innovative Technology and Exploring Engineering, 2021, 10(2): 105-110.

损失函数	方案	数据增强方式	测试准确率/%	损失值/%
Focal loss	F	普通增强	94.31	28.59
Focal loss	FX	Mixup混合增强	94.68	27.12
Cross entropy loss	C	普通增强	94.26	33.59
Cross entropy loss	CX	Mixup混合增强	94.57	32.98

模型	数据规模/MB	测试准确率/%	单张图像检测时间/s
迁移VGG16	98.38	86.62	0.57
迁移ResNet50	92.14	89.95	1.54
迁移Inception-V3	506.28	90.26	0.55
迁移MobileNetV3	43.57	94.68	0.39
迁移 + 改进MobileNetV3	46.73	98.25	0.27

标签	类别	改进前/%	改进后/%
0	细菌病	95.94	98.20
1	早疫病	95.83	98.19
2	晚疫病	97.97	99.21
3	叶霉病	89.95	97.70
4	七星叶斑病	96.28	99.22
5	靶斑病	96.07	98.02
6	双斑蜘蛛	90.90	97.21
7	花叶病毒	94.89	98.47
8	黄曲叶病	95.08	97.37
9	健康	93.92	98.93

[1]	ZHANG Zhibo, ZHAO Xining, GAO Xiaodong, ZHANG Li, YANG Menghao. Accurate Extraction of Apple Orchard on the Loess Plateau Based on Improved Linknet Network [J]. Smart Agriculture, 2022, 4(3): 95-107.
[2]	CHEN Zhanqi, ZHANG Yu'an, WANG Wenzhi, LI Dan, HE Jie, SONG Rende. Multiscale Feature Fusion Yak Face Recognition Algorithm Based on Transfer Learning [J]. Smart Agriculture, 2022, 4(2): 77-85.
[3]	WEI Jing, WANG Yuting, YUAN Huizhu, ZHANG Menglei, WANG Zhenying. Identification and Morphological Analysis of Adult Spodoptera Frugiperda and Its Close Related Species Using Deep Learning [J]. Smart Agriculture, 2020, 2(3): 75-85.
[4]	Li Miao, Wang Jingxian, Li Hualong, Hu Zelin, Yang XuanJiang, Huang Xiaoping, Zeng Weihui, Zhang Jian, Fang Sisi. Method for identifying crop disease based on CNN and transfer learning [J]. Smart Agriculture, 2019, 1(3): 46-55.
[5]	Chen Guifen, Zhao Shan, Cao Liying, Fu Siwei, Zhou Jiaxin. Corn plant disease recognition based on migration learning and convolutional neural network [J]. Smart Agriculture, 2019, 1(2): 34-44.